With the advancement of technology, media content have been migrated from analog to digital format. Although analog to digital transition has offered better user-experience and new means of content production, distribution, and monetization, it has made illegal reproduction and distribution of digital content easier. Piracy of digital media content is increasing day by day and is a major cause of worry for the digital content owners.
Rapid growth of technology has enhanced the usage and distribution of digital media. Although digital media is enriched with several advantages such as easy to store, transmit, etc., it has raised many security concerns such as protection of digital rights. Movie industry is incurring a loss of billions of dollars every year due to piracy and violation of digital rights. In order to prevent piracy and other digital rights violations, digital watermarking has emerged as a potential solution. Secret information is embedded inside the digital media which is later extracted to prove the ownership of the media element.
The protection of a digital content is a major concern nowadays. Earlier cryptography based methods are used for the protection of digital rights. But these traditional cryptography based methods fail to protect the digital content from copyright infringement. Digital watermarking has emerged as a solution to protect the digital content. Digital watermarking is a fundamental step towards Copyright Protection, Authenticity and Digital Rights Protection.
The ongoing distribution of digital media is managed by Digital Rights Management (DRM) based systems. DRM based systems ignore the Traditional Rights and Usages (TRU) of digital media users and hence, suffer from interoperability and mobility issues. To preserve the TRU of digital media users and combat piracy, digital watermarking based systems are proposed.
Video watermarking is an active area of research for last decade. A number of video watermarking algorithms are proposed by the researchers. These algorithms can be classified into two domains:
Pixel domain video watermarking—The watermark is embedded in the video frames by simple addition or bit replacement of selected pixels. These methods are computationally fast but less robust.
Transform domain video watermarking—The video frame is transformed and watermark is embedded in the transform coefficients. These methods are robust to common signal processing attacks like compression, etc. but require high computational time.
The video watermarking techniques should satisfy various characteristics such as Imperceptibility, Robustness, Real-time embedding, Security, Blind detection and Capacity. All of these characteristics are hard to attain in a single watermarking algorithm and hence, a trade-off is required between the characteristics. For example, transform based methods are robust to signal processing attacks but they require high computational time which makes them applicable only to stored videos and not for real-time applications.
The existing algorithms have limitations as most of the algorithms are based on transform domain due to their robustness against signal processing attacks. These algorithms are applicable only to the stored videos. The algorithms (robust) embed watermark either using the frequency domain transforms or variable length code words (VLC). These algorithms require high computational time.
Further, the existing algorithms are applicable to stored videos (compressed or uncompressed domain) and these algorithms are not capable of processing the video in real-time because of high computational time.
Thus, there is a need to overcome the problems of the existing technologies. Therefore, the present inventors have developed a computer-implemented method, system and computer readable medium for embedding a watermark into a video in real-time while capturing it or while delivering or streaming it i.e. on-the-fly and extracting the watermark from the watermarked video which would protect the videos from copyright infringement and authenticity.